The epigenomic modulation of gene function is well known. Cytosine bases (C) in DNA may be found methylated in the 5 position, and this is used as a signal to reduce expression of the genes in which those bases are located. Enzymes add methyl groups to C de novo to maintain the pattern of methylation through DNA replication.
Epigenetic processes control cell differentiation (allowing cells to maintain different characteristics despite containing identical genes); imprinting; gene silencing; X-chromosome inactivation; reprogramming; the progress of carcinogenesis; etc. Embryonic development, as well as success of cloning and embryonic stem cell (ESC) technologies, depend on the epigenetics. DNA methylation patterns are controlled by DNA methyltransferases, of which DNMT1 is the most abundant. It transfers patterns of methylation to a new strand after DNA replication and is essential for embryonic development, imprinting and X-inactivation [Robertson K D. et al, Nat. Rev. Genet. 2000; 1:11]. Epigenetic patterns “reset” when organisms reproduce.
In mammals, most cells terminally differentiate, and only stem cells retain the ability to differentiate into different cell types. Embryonic stem cells can differentiate into any cell type, thus holding a tremendous potential for regenerative medicine. As embryonic stem cells are difficult to obtain, methods of reprogramming other cells, such as skin cells, to give induced pluripotent stem (iPS) cells are being developed. Current approaches to trick cells into de-differentiation rely on genetic modification [Okita K. et al. Nature 2007; 448:313]. De-differentiation is based on changes in DNA methylation patterns, so novel methods of affecting the DNA methylation process and erasing the DNA methylation pattern are required. Any method that provides a bias on methylation pattern or its kinetics would be a valuable research tool.
Preferably, such methods should not involve genetic manipulations or immunogenic drugs, which currently represent a major disadvantage of the existing methods—such as immunogenicity, irreversibility of action, toxicity, instability under physiological conditions, etc. [Christman J K Oncogene 2002; 21:5483]. We propose a novel principle of modulating DNA methylation that may be free of these drawbacks, broadening the arsenal of epigenetics R&D tools.
Many diseases are related to changes in gene expression, and therapies, which alter patterns of gene expression, are useful for treating those diseases. Hence, treatments have been proposed which alter the pattern of DNA methylation through inhibiting the methylation or demethylation of DNA by affecting the enzymes concerned, as discussed for example by Yoo et al. (Biochem. Soc. Trans 32 (6): 910-912). The use of methylation inhibitors have been suggested and tested for the treatment of a wide range of diseases, including cancers such as bladder cancer (Zhang et al., Urologic Oncology—seminars and original investigations 24(2): 1520160), haematopoietic cancers (Jost et al., Letters in drug design and discovery 3(4): 242-252).
Epigenetics affects genetic diseases through imprinting, i.e. different epigenetic patterns contributed by the father and mother (Angelman, Prader-Willi syndromes, etc.). Some teratogens exert damage to the fetus by epigenetic mechanisms.
Gene silencing by aberrant methylation of promoter regions of genes critical for normal cellular functions is a hallmark of cancer [Jones P A. et al.; Nat. Rev. Genet. 2002; 3:415], although a small number of genes (about 10%) in cancerous cells are actually hypomethylated compared to normal cells [Ehrich M. et al., PNAS 2008; 105:4844]. Epigenetic carcinogenes (hexachlorobenzene, arsenite), while not being mutagenic, still result in an increased incidence of tumors. Drugs have been developed that affect DNA methylation by inhibiting corresponding enzymes [Yoo C B. et al., Bloch. Soc. Trans. 2004; 32, 910; and refs therein]. This approach has been tested against a wide range of diseases, including cancers such as bladder cancer, haematopoietic cancers, etc. [Jost E. et al., Letters in Drug Design & Discovery 2006; 3:242].
Cytidine analogues unmethylatable at position 5, such as 5-fluoro-cytidine or 5-azacytidine have been suggested as drugs for cancer therapy (for example WO2004050666) and specific analogues which might modulate DNA methylation are disclosed in, WO2006099132, US2006/0205687, and are reviewed by Jones and Taylor (Cell 20:85-93). A-azazytidine has been approved by the U.S. Food and Drug Administration and is being used as a treatment for cancer under the name Decitabine and for the treatment of Myelodysplastic syndrome under the name Vidaza. These analogues all have a different chemical structure from the cytosine present in normal tissues. Other agents that are not cytosine analogues have also been disclosed which alter methylation of DNA, such as those disclosed in WO2005/011661 and WO2005/085196.
A drawback of all such treatments is that they can act to completely block or overwhelmingly enhance DNA methylation. They are therefore likely to have adverse effects on genes other than the target gene whose activity is desired to be modulated, but not necessarily eliminated (as reviewed by Haaf et al., Pharmac. Therap. 65:19-46). The harmful effects can sometimes be avoided by careful adjustment of the dose of the inhibiting drug, but this requires skill, cognizance of patient physiology and variability, and is not always possible.
It is known that the rates of certain chemical reactions are affected by the nature of the isotopes of the atoms in the reacting bonds. In general, bonds terminating in a heavy isotope will be less liable to cleavage than a bond terminating in a lighter isotope. Of particular note is that bonds between hydrogen atoms and other atoms are less liable to breakage if the hydrogen is 2H rather than 1H. A similar effect is seen when comparing the rate of cleavage of a bond between a carbon atom and another atom, where bonds with 13C are less liable to cleavage than bonds with 12C. This is known as the Kinetic Isotope Effect, and is well described. Many isotopes are known to show this effect, as is described in Isotope effects in chemical reactions. (C. J. Collins, N. S. Bowman (eds.) 1970). It is known that these effects are also manifest in enzyme-catalysed reactions, as described in Isotope effects on enzyme-catalysed reactions (Cleland, W. W., M. H. O'Leary, and D. B. Northrop (eds.) 1976).
This invention arises from understanding that the inherent drawbacks of methylation modulating drugs may be overcome by using an agent which modulates but which 1) is naturally occurring and therefore not toxic or immunologic, and 2) can partially inhibit the methylation or demethylation of C in controlled ways. This invention describes the use of the kinetic isotope effect to achieve this effect, and its potential use as both a research tool and possible therapy.